High-resolution LA-ICP-MS trace element mapping of igneous minerals: In search of magma histories

We report experiments on optimisation of LA-ICP-MS mapping as a tool for visualising and quantifying internal structure of trace element concentration in igneous minerals. The experimental design was refined with maps on clinopyroxene and amphibole macrocrysts (mainly antecrysts) from a porphyritic lamprophyre in NE Spain, as well as on a high precision metal wire grid. In terms of spatial resolution, we demonstrate with scanning electron microscope and white light interferometry that a full ablation removes between 0.4 and 0.7 mu m of material, depending on ablation parameters. Maps were produced with square laser beam spots of 12 and 24 mu m. It was found that complexities can be resolved in the sample even though they are smaller than the beam diameter (e.g., 7-10 mu m discontinuities using 12 mu m laser beam). Resolution in x and y was found to be identical, probably reflecting the fast washout of the two-volume ablation cell and the short total dwell time of the analyte menu selected. Due to the excellent stage reproducibility and the limited ablation depth, it is feasible to re-ablate the identical map area many times employing different instrument parameters or analyte menus. On the magmatic crystals, LA-ICP-MS maps define very sharp compositional zoning in trace elements, highlighting complex crystallisation histories where 'normal' magmatic fractionation is not the only process. Events of mafic recharge are easily recognised as zones enriched in compatible metals such as Cr, Ni or Sc. Further, trace element maps reveal complexities in mineral zoning previously undetectable with petrography or major element data. These include resorbed primitive cores and oscillatory zoning within apparently homogeneous mineral zones. Therefore, LA-ICP-MS mapping opens a new window of opportunity for analysis of magmatic histories. The wide combination of instrumental parameters, such as laser beam size, scan speed and repetition rate, make it possible to carry out experiments at different levels of detail. We recommend a two-step approach to mapping. The initial step involves rapid maps to gain an overview of potential complexities in the sample; this enhances representativeness of the analysed materials, as a large number of crystals and trace elements can be tested in little time. Subsequently, detailed maps can be carried out on areas of interest. An additional functionality is to create 1D-profiles from 2D-maps. The potential of the technique to unveil compositional complexities efficiently and at greater detail than traditional microanalysis will help to improve our understanding of processes in the magmatic environment and beyond. (C) 2015 Elsevier B.V. All rights reserved